Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Crearize/ai-dev-helm --skill subagent-driven-developmentgit clone --depth 1 https://github.com/Crearize/ai-dev-helmWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/crearize/ai-dev-helm/subagent-driven-development)<a href="https://agentmods.dev/skills/crearize/ai-dev-helm/subagent-driven-development"><img src="https://agentmods.dev/badge/skills/crearize/ai-dev-helm/subagent-driven-development/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/crearize/ai-dev-helm/subagent-driven-development"><img src="https://agentmods.dev/badge/skills/crearize/ai-dev-helm/subagent-driven-development.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00015 | $0.00433 |
| Opus 5.5 | $0.00006 | $0.00173 |
| Sonnet 5.5 | $0.00003 | $0.00087 |
| Haiku 4.5 | $0.00002 | $0.00043 |
Grade A, and why
subagent-driven-development scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
100% identical to subagent-driven-development — 564 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
Subagent-Driven Development
Choose delegation by total effort: briefing, implementation, rediscovery, integration, and rework. The parent may implement or fix code directly. Do not require a particular implementation model or forbid the parent model from writing code. Select available models for task difficulty and use the runtime's actual tool contract.
Delegate only work with clear ownership, interfaces, acceptance criteria, and useful independence. Parallel tasks must not edit the same files or depend on unfinished shared state. Give each worker concise requirements, relevant source paths, the workspace, constraints, and expected verification; do not ask it to transcribe a complete implementation from a plan.
Use implementer-prompt.md as an optional brief template. The controller integrates results, resolves dependencies, and can handle a small fix itself. Workers report evidence and uncertainties; they do not launch extra reviewers. Maintain a concise progress record only when it aids resumption, using existing runtime records where available.
Run appropriate tests while implementing. After integration, use the project's independent quality-check and its review budget. There is no mandatory task-by-task review, spec/quality double review, or additional whole-branch reviewer. A targeted investigation can still be requested when a concrete unresolved concern requires another perspective; account for it in the existing review budget.
The task-brief and workspace scripts remain optional utilities for larger handoffs. Legacy task review/re-review templates are optional formats for an explicitly selected quality-check investigation, not additional gates. Do not create their ledgers, snapshots, or packages unless the chosen workflow needs them.
Workers report changed files, test evidence, deviations from the design and limitations to the controller; the controller records the deviations. Before completion, verify the integrated result; the report to the user follows the final-message rule in documents/development/development-policy.md §1.0 "承認後の進め方" (deviations from the design only; test evidence goes into the PR body). Continue integration within existing authorization and project branch policy.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago Changed 8c96bc56a8d0
- 1mo ago Changed · -550 lines · -2 tokens per session scan C → A 8965b40ebcdb
- 1mo ago First seen · 569 lines · 17 tokens per session scan C 8dd1b8e698ed
subagent-driven-development is a skill published in the GitHub repository Crearize/ai-dev-helm (5 stars, last pushed 4d ago), licensed MIT. It adds 15 tokens to every session and 433 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 100% identical to subagent-driven-development, differing in 564 lines, and is treated as a copy.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…